Difference between estimates from SAS Proc Genmod and Raw averages -

i trying find average change in score survey response pre post @ different levels of aggregation using 2 different methods. in method 1 called "raw method", calculate average of pre scores , average of post scores , take difference. in method 2 " genmod method", use pre , post categorical variable , estimate of interaction term (stage2*rso). why not estimate method 2 match average change in scores in method 1? dataset has multiple pre's , posts each patient (empi). simple problem need understanding. heres dataset , code:

  empi  provider    rso score   stage   1001    x   16.5    pre   1001    x   22.2    post   1001    x   14.3    pre   1001    x   23.4    post   111     x   25.6    pre   1002  b   x   32.3    pre   1002  b   x   12  post   1001    x   24.3    pre   1002  b   x   15.6    pre   1002  b   x   23.7    post   112     x   10.2    pre   1234  c   y   13.5    pre   1234  c   y   34.2    post   1234  c   y   12.3    pre    /** method 1 **/   proc sql;    create table rsoscores     select         stage2,             rso,             avg(score) avg        test1       group stage2,               rso;     quit;    /** method 2 **/    proc genmod data=test1;    class empi rso stage2;    model score  =  stage2 rso stage2*rso;    repeated subject = empi/ type=un corrw;   ods output geeemppest = mygee_pes;    run;